The long term goals of this research are to produce accurate, quantitative assessments of ultrasound scattering in vivo, to improve the noninvasive diagnostic capabilities of medical ultrasound. This research will center on the diagnosis and monitoring of diffuse liver disease with ultrasound and the detection and differentiation of focal lesions, especially when these are masked by diffuse disease conditions. The hypothesis is that quantitative backscatter images depicting backscatter levels and backscatter frequency dependence, as well as acoustic parameters derived from backscatter signals, including attenuation, scatterer size estimated from spatial auto correlation functions and scatterer number densities, will provide improved differentiation between normal and diseased conditions. Instrumentation will be adopted to expand the frequency range of liver backscatter estimates, and a data base on backscatter and attenuation coefficients for normal individuals will be expanded. Newly derived parameters for assessing scatterer size and number density will be further evaluated in phantoms, including novel ones constructed to model liver disease. When thoroughly understood, values for these parameters will be obtained for normal liver. Work will continue to measure backscatter, attenuation, and when available, scatterer size and number density in a large group of patients with diffuse liver disease. These studies will be extended to individuals with proven focal lesions in the liver, exploring new ways to acquire and display echo data. Finally, animals models for fatty infiltration in the liver, cirrhosis and steroid induced hepatophy will be used to further our understanding of acoustic changes accompanying liver disease.